1. The Choices Governors Make: Public Goods and Corruption in the Brazilian states Marcus Melo (Universidade Federal de Pernambuco) Lee Alston (University of Colorado) Bernardo Mueller (Universidade de Brasília) Carlos Pereira (Michigan State University/FGV-SP)

2. Motivational Puzzle: Brazilian states exhibit great similarity with respect to their macro level institutional features: Politicians are elected every four years with the same electoral rules Governors are allowed to run for re-election just once and are very powerful, equipped with several institutional tools to govern; the decision-making process within state legislatures is very centralized with a extremely weak and unprofessional committee system; in fact, legislative bodies are mostly reactive to executive dominance; the state courts are formally independent and sometimes work as an important constraint to the executives preferences; every state possesses accountability institutions to deal specifically with the propriety of government expenditures (Tribunais de Contas)

3. Motivational Puzzle: Even with those great similarities in their institutional endowments, the Brazilian twenty-seven sub-national unities are very distinct with regard to their economic and policy outcomes. It is possible, on the one hand, to find states with a great level of social and economic development, and on the other, states that are extremely poor. If the macro state institutional endowments do not present enough variation to explaining these different outcomes, what are the other institutional and political aspects that can account for these differences?

4. Research questions What factors account for the wide variation in developmental policy outcomes in the Brazilian states? Why some states exhibit good capacity for policy coordination and adaptability whereas other states are characterized by policy inertia retaining inappropriate policies for long periods? Why some state governments are able to provide public goods such as a professionalized bureaucracy, fiscal balance and adequate health and education services?

5. Inefficient predatory. Do societies choose inefficient policies and institutions? This paper … develops the argument that there are strong empirical and theoretical grounds for believing that inefficient policies and institutions are prevalent. We conclude that these inefficient institutions and policies are chosen because they serve the interests of politicians or social groups that hold political power at the expense of the rest. Put even more strongly ; why do powerful groups not predate efficiently ? (Acemoglu, 2002)

6. Policy Failure and Political Survival: The Role of Political Insitutions (Bueno de Mesquita et al., 1999) It seems obvious

7. Political Competition and Economic Performance: Theory and Evidence from the United States Besley and Persson, 2005. One of the most cherished propositions in economics is that, by and large, monopoly is bad and market competition between firms raises the welfare of consumers. Whether competition between political parties has similarly virtuous consequences is far less discussed, despite the long-term monopoly on power by a dominant party observed in a number of existing democracies. Moreover, almost no empirical studies speak to the question if political competition matters at all for economic outcomes.

8. Our Claim: Other economic aspects such as the stock of investment, level of economic integration with other state unities and with the international market, foreign investment, etc., play an important role on economic and political outcome. However, we stress that other micro institutional aspects related to the state politics and policymaking play a key role in explaining different economic and political performance at the sub-national level in Brazil

9. Contestability HighLow Checks & Balances Low Predatory political environments (politicians with short policy horizons) Ex. Rondônia Patrimonialist entrepreneurial politics Ex. Bahia High Governance-enhancing incentives (may produce policy volatility if preferences are polarized) Ex. Rio Grand do Sul Governance- enhancing incentives Ex. Minas Gerais

10. Model of Governors Choice: Allocation of public resources: All society (public goods); Interest groups (private goods); own pocket (corruption). The governor of a state maximizes votes and money. The governor chooses: the amount of effort to produce public goods (E u ) the amount of effort to produce private goods (E R ) how much of the resources received from private groups are allocated to pursue reelection (α) and how much is pocketed for personal gain (1-α). There is a limited amount of effort available to the governor.

11. Governors utility is affected by: Subject to E u + E r = and 0 α 1. votes Public Goods Private Goods Resources

12. Model of Governors Choice: The first order conditions that solve this problem are: Where is the Lagrange multiplier on the restriction E u + E r =. This condition states that the marginal unit of effort will always be placed in that activity (public or private good) which yields the greatest electoral return to the governor, given α. Similarly, this condition states that the decision whether to use resources for electoral or for personal purposes is taken so that the marginal real (R$) goes to that purpose which generates most utility. Thus in equilibrium, the utility from the marginal real is the same whether it goes to finance the governors campaign or his bank account.

13. The equilibrium values of the dependent variables E u, E r, and α depend on a series of parameters that affect the various functions. These parameters capture the characteristics of each individual states institutions, special circumstances and idiosyncrasies: Comparative static Productivity of effort in producing public goods: (·, ) Productivity of effort in producing private good: (·, ) The electoral response to public goods: (·, ) The electoral response to private goods: (·, ) The marginal utility of votes to the governor: U V (·, ) The productivity of private policies in generating resources: (·, ) The voters sensitivity to electoral campaigns : V R (·, ) The marginal utility of money to governador: U M (·, ) Parameters

14. Theoretical Framework These productivities in turn depend on: Political competition (contestability) Effective number of parties; Electoral competition; Campaign contributions; etc. Institutionalization Rule of Law; Checks and balances; Veto points; etc. Other factors Education; GDP; Gini coefficient etc.

15. Dependent variables (Public vs. Private goods): An index of expenditure efficiency (variation and average) in the states (Ferreira Júnior, 2006) The index is a ratio of the part of total expenditure that is effectively spent in the final public good that is being provided (including debt) divided by the administrative and other intermediary costs involved in producing those services. States with a higher value of this index provide more public goods at a lower cost. Expenditure by state governments on programs of voluntary retirement. these programs have investment-like qualities, where the government pays the quitting civil servant an upfront compensation and receives only long term benefits as its salary and pension bill diminishes over time.

16. Dependent variables: Corruption? Average of the percent of wealth variation for state deputies In order to provide a measure that proxy for the amount of personal benefit the governors and other politicians achieve from office, we use data from the Superior Electoral Tribunal that requires all candidates to political office to publicly declare their wealth. Average of the percent of wealth variation for all politicians in a particular state

17. NumState Regulatory AgenciesJudiciary Public Prosecutors Audit OfficeCNJMedia Public Defenders 1Acre591121522 2Alagoas1910256516 3Amapá62422121012 4Amazonas218313 24 5Bahia26752711723 6Ceará418621996 7Distrito Federal7237263249 8Espírito Santo812818162513 9Goiás25139238418 10Maranhão91102431 11Mato Grosso2014131417 14 12Mato Grosso do Sul24111210152627 13Minas Gerais10511823194 14Pará113143261611 15Paraíba22415177117 16Paraná1326187251419 17Pernambuco22171625182310 18Piauí1219174182 19Rio de Janeiro2715191619203 Rio Grande Norte125201121226 21Rio Grande Sul23222324 2725 22Rondônia1421 620185 23Roraima15622125117 24Santa Catarina1627249272120 25São Paulo34261522 21 26Sergipe1716252014128 27Tocantins1820271910615 Inputs in the Checks & Balances Index

18. Checks & Balances Index Ranking NumStateIndex 1Rio Grande do Sul0.89 2Santa Catarina0.76 3Pernambuco0.69 4Mato Grosso do Sul0.66 5Paraná0.65 6Rio de Janeiro0.63 7Tocantins0.61 8São Paulo0.60 9Sergipe0.59 10Mato Grosso0.58 11Bahia0.56 12Rio Grande do Norte0.56 13Rondônia0.56 14Espírito Santo0.53 15Goiás0.53 16Distrito Federal0.52 17Amazonas0.50 18Pará0.44 19Paraíba0.44 20Minas Gerais0.42 21Roraima0.41 22Ceará0.39 23Amapá0.36 24Alagoas0.33 25Piauí0.33 26Acre0.29 27Maranhão0.16 Mean0.52 Std. Dev.0.16

19. Institutionalization X GDP per capita (2003)

20. Determinants of Politicians Wealth Variation (1) State Deputies Wealth Variation (2) All Politicians Wealth Variation Checks & Balances index -3.736 * (-1.66) -1.035 ** (-2.32) Electoral competition State Assembly -0.477 *** (-2.81) -0.095 * (-1.70) # effective parties House of Repres. 0.512 *** (2.83) Electoral competition House of Represent. 0.135 (0.88) # Parties in Coalition in State Assembly -0.143 ** (-2.31) Expected margin for reelected governors 0.440 * (1.87) Lame duck Governor 0.536 (1.23) 0.249 (1.39) Pork 0.002 ** (2.40) Education 0.634 (1.57) -0.016 (-1.35) N 0 state deputies in wealth var. variable -0.050 * (-1.81) Constant 1.54 (0.85) 1.164 ** (2.04) Observations5427 R-squared (adjusted) within: 0.50 between: 0.46 overall: 0.48 0.31 (0.10)

21. Determinants of Expenditure Efficiency (1) Expenditure Efficiency (Variation) (2) Expenditure Efficiency (Average level) (3) Voluntary Retirement Program Checks & Balances 1.230 ** (2.10) 1.441 * (1.72) 18.728 *** (2.67) Initial level of Expendit. Efficiency -0.090 (-1.41) Electoral competition House of Represent. -0.158 * (-1.70) -0.389 ** (-2.73) Electoral competition State Assembly 0.570 *** (2.99) 0.240 *** (3.55) 3.298 *** (3.54) Electoral competition State Assembly squ. -0.044 *** (-2.88) Effective number parties State Assem. -0.0810 ** (-2.27) Party concentration in State Assembly 16.122 * (1.91)

22. (1) Expenditure Efficiency (Variation) (2) Expenditure Efficiency (Average level) (3) Voluntary Retirement Program Number of Parties in Gov.s Coalition 0.037 (1.21) Margin of victory in gubernatorial election -0.147 (1.08) 0.964 (1.25) Lame duck Governor -0.285 ** (-1.98) Pork (per gdp) 0.077 *** (4.55) Gini -151.21 *** (2.53) 13.71 *** (3.47) Gini squared 145.07 *** (2.67) Education 0.287 ** (2.32) Constant 38.44 ** (2.36) -6.167 (-2.53) -34.399 *** (-3.53) Observations54 27 R-squared (adjusted) within: 0.28 between: 0.70 overall: 0.48 within: 0.19 between: 0.59 overall: 0.42 0.90 (0.87)

23. Number of Parties in the Governors Coalition x C&B in Wealth Variation Equation

24. Interaction of the Effective N o of Parties in the House and C&B

25. Interaction of Electoral Competition in the State Assembly and C&B

26. Interaction of Expected Governors with Long-Term Horizons and C&B

27. Conclusion checks & balances and political competition are major determinants of the decisions of governors, a result which should not come as a surprise but for which there is not a wealth of empirical evidence in the literature. Better developed checks & balances have a strong impact on the choices of governors- advancing public goods and restricting the provision of private goods and the pursuit of personal benefits.